GABAB Presynaptic Inhibition Has an In Vivo Time Constant

advertisement
J Neurophysiol
87: 1196 –1205, 2002; 10.1152/jn.00077.2001.
GABAB Presynaptic Inhibition Has an In Vivo Time Constant
Sufficiently Rapid to Allow Modulation at Theta Frequency
BRADLEY J. MOLYNEAUX AND MICHAEL E. HASSELMO
Department of Psychology, Program in Neuroscience, and Center for BioDynamics, Boston University, Boston,
Massachusetts 02215
Received 29 January 2001; accepted in final form 31 October 2001
Molyneaux, Bradley J. and Michael E. Hasselmo. GABAB presynaptic inhibition has an in vivo time constant sufficiently rapid to allow
modulation at theta frequency. J Neurophysiol 87: 1196 –1205, 2002;
10.1152/jn.00077.2001. Cyclical activity of GABAergic interneurons
during theta rhythm could mediate phasic changes in strength of
glutamatergic synaptic transmission in the hippocampal formation if
presynaptic inhibition from activation of GABAB receptors is sufficiently rapid to change within a theta cycle. The experiments described here analyzed the time course of GABAB modulation using a
heterosynaptic depression paradigm in anesthetized rats at physiological temperatures. Heterosynaptic depression of the slope of evoked
potentials decayed with a time constant that would allow significant
changes in transmission across different phases of the theta cycle. This
heterosynaptic depression was significantly reduced by local infusion
of the GABAB receptor antagonist CGP55845A.
INTRODUCTION
Theta rhythm oscillations appear in the hippocampal electroencephalogram (EEG) during active exploration (Buzsaki et
al. 1983; Chrobak and Buzsaki 1994; Fox et al. 1986; Green
and Arduini 1954) and under urethan anesthesia (Fox et al.
1986). These oscillations are accompanied by cyclical changes
in the slope of evoked excitatory synaptic potentials (Rao et al.
1998; Wyble et al. 1997, 2000) and population spike amplitude
(Buzsaki et al. 1981; Rudell and Fox 1984; Rudell et al. 1980).
These changes could result from rhythmic inhibition from the
medial septum (Stewart and Fox 1990; Toth et al. 1997)
causing phasic changes in interneuron activity during theta
(Fox et al. 1986; Skaggs et al. 1996).
Phasic interneuron activity could cause cyclical changes in
the magnitude of synaptic transmission. Data from slice preparations demonstrate that activation of presynaptic GABAB
receptors inhibits glutamatergic transmission in stratum radiatum at synapses between pyramidal cells within the hippocampus (Ault and Nadler 1982; Colbert and Levy 1992; Vogt and
Regehr 2001). In contrast, GABAB agonists do not inhibit
transmission at perforant path synapses in s. lacunosum-moleculare, which bring afferent input from entorhinal cortex
(Ault and Nadler 1982; Colbert and Levy 1992). The selective
presynaptic inhibition of internal connections could allow new
sensory input arriving via afferent fibers to dominate. This
would provide effective dynamics for encoding of new information due to synaptic modification of synapses in s. radiatum,
Address for reprint requests: M. E. Hasselmo, Dept. of Psychology, 64
Cummington St., Boston, MA 02215 (E-mail: hasselmo@bu.edu).
1196
without interference from previous memories due to synaptic
transmission at previously modified synapses (Hasselmo et al.
2002; Sohal and Hasselmo 1998a,b; Wallenstein and Hasselmo
1997). Subsequently, rapid decay of this presynaptic inhibition
could allow a dominant influence of intrinsic connections to
mediate retrieval of previous information. Oscillations between
encoding and retrieval dynamics within each cycle of the theta
rhythm would require a time course of GABAB presynaptic
inhibition sufficiently rapid to change within each theta cycle.
One method of measuring the time course of GABAB presynaptic inhibition of glutamatergic transmission is to analyze
heterosynaptic depression (Isaacson et al. 1993). The putative
mechanism for heterosynaptic depression is summarized in
Fig. 1A. In this experimental phenomenon, strong “conditioning” stimulation of one synaptic input to a region causes
spiking of local inhibitory interneurons which release GABA.
The GABA diffuses to presynaptic receptors on glutamatergic
terminals and suppresses excitatory synaptic potentials elicited
by subsequent “test” stimulation through a different stimulating electrode (Dittman and Regehr 1997; Isaacson et al. 1993;
Vogt and Nicoll 1999). The heterosynaptic depression elicited
in these conditions is blocked by GABAB receptor antagonists,
ruling out a role of other receptors. The heterosynaptic depression is accompanied by an increase in paired-pulse facilitation.
The increase in paired-pulse facilitation is taken as evidence
for presynaptic inhibition (Dittman and Regehr 1997; Valentino and Dingledine 1981; Zucker 1989) because postsynaptic inhibition should affect both potentials similarly, whereas a
decrease in release of transmitter induced by the initial pulse
should increase the available store for the second pulse.
Previous work has analyzed the time course of this effect in
slice preparations with different delays between conditioning
and test stimulation, showing a peak at about 300 ms and a
return to baseline around 1.4 s in slice preparations of hippocampal region CA1 (Isaacson et al. 1993). A similar peak is
seen in cerebellar slices (Dittman and Regehr 1997). Depression of inhibitory synaptic potentials by activation of presynaptic GABAB autoreceptors shows a similar time course in
hippocampal slices (Otis et al. 1993). These time courses were
obtained at nonphysiological temperatures (24 and 32°C in
Dittman and Regehr 1997, 30 –32°C in Isaacson et al. 1993),
but the Q10 derived for this effect (Dittman and Regehr 1997)
The costs of publication of this article were defrayed in part by the payment
of page charges. The article must therefore be hereby marked ‘‘advertisement’’
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
0022-3077/02 $5.00 Copyright © 2002 The American Physiological Society
www.jn.org
TIME COURSE OF GABAB PRESYNAPTIC INHIBITION
1197
FIG. 1. A: schematic circuit illustrating the theoretical mechanism of heterosynaptic depression in physiological studies of
hippocampal region CA1. The circle is a schematic representation of a region CA1 pyramidal cell receiving synaptic input
from fibers activated with either ipsilateral stimulation (left) or
contralateral stimulation (right). Ipsilateral stimulation causes
activation of local inhibitory interneurons (represented schematically by small circle). These local interneurons release GABA,
which activates presynaptic GABAB receptors on the terminals
of fibers activated by contralateral stimulation. The activation of
presynaptic GABAB receptors inhibits the release of glutamate
from the terminals of fibers activated by contralateral stimulation, thereby decreasing the rising slope of these evoked
potentials. (The heterosynaptic depression is independent of the
excitatory synaptic potentials induced by the ipsilateral stimulation.) B: design of experiment. A recording electrode is positioned in s. radiatum of region CA1 on one side of the brain. An
ipsilateral stimulating electrode is positioned in region CA3 of
the same side, and a contralateral stimulating electrode is positioned in region CA3 of the opposite side. Evoked potentials
elicited with contralateral stimulation are recorded, allowing
comparison of the initial slope of evoked potentials before and
after an ipsilateral stimulus pulse.
suggests that the time course at physiological temperatures
might be much more rapid. Suppression of synaptic transmission has also been tested with rapid pulsed perfusion of baclofen (Pfrieger et al. 1994) or photolysis of caged compounds
(Dittman and Regehr 1997). However, these experiments were
also performed at nonphysiological temperatures.
To obtain a measurement of the time course of presynaptic
GABAB receptor effects in normal physiological conditions,
the time course of heterosynaptic depression was tested in vivo
in anesthetized rats at physiological temperatures. This allowed
measurement of the faster time course with natural receptor
and second-messenger kinetics and functional GABA reuptake
mechanisms.
Surgery
Male Sprague-Dawley rats (325–500 g) obtained from Charles
River (Wilmington, MA) were anesthetized with an intraperitoneal
injection of urethan (1.5 g/kg), and anesthesia was maintained with
additional injections of urethan as necessary. Body temperature was
monitored with a rectal thermometer and maintained at 37°C (Lillie et
al. 1996) with a heating pad. Rats were placed in a David Kopf
stereotaxic frame, and holes were drilled into the skull using a David
Kopf stereotaxic drill. The recording electrode and stimulation electrodes were guided stereotaxically through previously drilled holes in
the skull according to coordinates from Paxinos and Watson (1986).
The recording electrode was placed in the s. radiatum of region CA1
METHODS
To test heterosynaptic paired-pulse depression in vivo, two
separate pathways into s. radiatum of hippocampal region CA1
were stimulated during extracellular recording of evoked synaptic
potentials in this region, allowing recording of the excitatory
synaptic transmission from these pathways as negative-going field
potentials. The location of stimulation and recording electrodes is
summarized in Fig. 1B. Release of GABA was elicited by a single
pulse or a pair of pulses through a stimulating electrode in region
CA3 ipsilateral to the recording electrode. Subsequently, two stimuli were presented through the contralateral stimulating electrode
at intervals ranging between 30 and 1,500 ms after the last ipsilateral pulse. The magnitude of the initial negative slope of these
stimuli was compared with baseline evoked potentials obtained
with contralateral stimulation 10 s preceding the ipsilateral stimulation. Pairs of stimuli were used at the contralateral stimulation
site to test for changes in the magnitude of paired pulse facilitation.
The relative timing of contralateral and ipsilateral stimulation
pulses is summarized in Fig. 2.
J Neurophysiol • VOL
FIG. 2. Timing of synaptic stimulation during experiment. Each trial of the
experiment consisted of the following stimulation. First, a pair of pulses (—)
are delivered through the contralateral stimulation electrode. These are used to
establish a baseline magnitude of evoked potential slope (examples of baseline
evoked potentials are shown). This is followed by a 10-s delay to ensure the
baseline does not play a role in heterosynaptic depression. Second, a single
stimulus pulse (- - -) is delivered through the ipsilateral stimulation electrode to
cause putative local release of GABA. This is followed by a variable delay of
30 –1,500 ms. Third, a pair of pulses (—) are delivered through the contralateral stimulation electrode. These are used to measure the change in rising slope
of evoked synaptic potentials caused by the previous ipsilateral stimulation.
The evoked potentials shown directly beneath the timeline show examples of
the evoked potentials elicited with each stimulus pulse. The potentials evoked
by contralateral stimulation show strong heterosynaptic depression of rising
slope in these potentials occurring at 40 ms after ipsilateral stimulation. The
use of a pulse pair allows analysis of the change in magnitude of paired-pulse
facilitation.
87 • MARCH 2002 •
www.jn.org
1198
B. J. MOLYNEAUX AND M. E. HASSELMO
(⫺3.1 mm from Bregma, 2.0 mm lateral, 2.5–3.0 mm deep). Stimulating electrodes were placed bilaterally in region CA3 (⫺3.8 mm
from Bregma, 3.7 mm lateral, 3.0 mm deep). Note that positioning of
stimulating electrodes required analysis of the evoked potentials elicited at the recording site with an emphasis on positioning electrodes so
that an ipsilateral pulse followed by a contralateral pulse did not
produce any paired-pulse facilitation, whereas two pulses to one
pathway produced paired-pulse facilitation (which appears to be primarily an intracellular, homosynaptic phenomenon).
Electrodes were custom-made stainless steel wires coated with
Formvar of 0.004-in-diam cut square (stimulating electrodes consisted
of 2 wires twisted together and recording electrodes consisted of a
single wire). Recording electrodes were connected to Grass P15
preamplifiers that fed into Neuralynx differential amplifiers connected
to a Data Translation A/D board for data storage and visualization in
a 100-MHz Pentium computer. All data acquisition was performed
using Experimenter’s Workbench Software by DataWave. Stimulation was controlled by a Grass S88 stimulator in conjunction with
Grass PSIU6 photoelectric stimulus isolation units.
Drug infusion
Administration of GABAergic antagonists was performed by local
infusion through a 25-gauge steel cannula (Small Parts) connected to
a 10 ␮l Hamilton syringe with fused silica tubing (Harvard Apparatus). The beveled cannula was fixed 1 mm from the recording electrode. Pharmacological agents were infused in 3 ␮l of artificial cerebrospinal fluid (ACSF) maintaining a flow rate of 0.2 ␮l/min. for 15
min with a microinfusion syringe pump (WPI). All drugs were dissolved in ACSF [which contained (in mM) 120 NaCl, 3.3 KCl, 25
NaHCO3, 1.33 NaH2PO4, 0.9 MgSO4, 1.3 CaCl2, and 10 dextrose].
CGP 55845A was infused at a concentration of 5 mM and 6-cyano7-nitroquinoxalene-2,3-dione (CNQX) at 1 mM. CGP 55845A (Davies et al. 1993) was a gift of Novartis AG (Basel). CNQX and all
other chemicals were purchased from Sigma.
Data analysis
All data analysis was performed with MATLAB version 5.2 (Mathworks) as in previous studies (Linster et al. 1999; Wyble et al. 2000).
For analysis, synaptic potentials were averaged across the five potentials obtained at a particular time interval between ipsilateral and
contralateral stimulation. The maximum magnitude of the initial negative slope of the averaged potential was then determined using a
sliding window of 0.42-ms duration (30 data points). This slope was
then normalized to a percentage value relative to the average slope of
the baseline potential recorded 10 s before the ipsilateral pulse in each
trial. Normalized slopes were then plotted according to the interval
between ipsilateral and contralateral stimulation and could be combined with data from other animals (with an error bar representing the
standard error resulting from variance across animals). The time
course of this depression was estimated by curve fitting of a dual
exponential curve to this time course. Homosynaptic facilitation was
measured by comparing response to the second pulse of the ipsilateral
paired-pulse stimulation with the corresponding first pulse at that
interval (i.e., the 2nd pulse of a pair at 10 and 50 ms was compared
with the 1st pulse of a pair at 50 and 90 ms), producing a plot of the
time course of change in paired-pulse facilitation. Paired-pulse stimuli
on the same pathway were presented with a 40-ms interval between
each pulse in the pair. In contrast, the interval between ipsilateral and
contralateral stimulation in testing heterosynaptic depression was varied between 30 and 1,500 ms.
To ensure there was not a systematic change in the time course of
depression, the delays between the conditioning and test stimulus
were tested using interdigitated blocks of intervals (e.g., 10, 40, 70,
J Neurophysiol • VOL
120 . . . followed by 20, 50, 80, 140 . . . followed by 30, 60, 100,
160 . . .). Comparison of interdigitated curves provided a measure of
any long-term drift in measurement of synaptic potentials, such as
would be produced by long-term depression or potentiation. A slow
increase in the magnitude of the negative slope of evoked potentials
over time was observed in some of the preparations. This was minimized by waiting 1 h after the final placement of the electrodes and
was controlled for by always calculating the percent depression using
a control potential evoked 10 s before the test potential. The increase
in negative slope over time was not due to long-term potentiation but
rather was most likely a result of recovery from the trauma of
electrode placement and surgery (Rick and Milgram 1999). Changes
in evoked potentials can be measured most effectively when these
potentials are not saturated. To avoid saturation, initial measurement
of the evoked potential was performed at a number of different
stimulus magnitudes to determine the asymptotic maximum of the
evoked potential amplitude. After determination of this maximum
amplitude, the stimulation was reduced until the evoked potential
amplitude was approximately 50% of the maximum amplitude, ensuring that the evoked potentials were recorded in the more central
linear range of the input-output curve. At the start of the experiment,
these evoked potentials had an initial amplitude of 0.8 –2.8 mV. The
use of interdigitated stimulation protocols would ensure that any
nonlinearities of response properties should cause inconsistencies in
the time course of heterosynaptic depression that were not detected.
There were differences in the time course of heterosynaptic depression and the change in paired-pulse facilitation depending on the
initial magnitude of paired-pulse facilitation. Therefore in the analysis, the experimental data were divided on the basis of a single
objective criterion based on the initial magnitude of paired-pulse
facilitation in the evoked potentials elicited by contralateral stimulation. If the contralateral paired-pulse facilitation was greater than
180% of control, the rats were grouped into the strong initial paired
pulse facilitation category. If this facilitation was less than 150% of
control, the rats were grouped into the weak initial facilitation category. These widely separate facilitation levels were chosen to reflect
the strong clustering of data at values below 150% of control [there
were 4 experiments in this category, with a mean facilitation of 145 ⫾
6% of control and (mean ⫾ SD)] and clustering of values around
200% of control (there were 7 experiments in this category with a
mean of 204 ⫾ 22%). Three experiments did not satisfy these criteria.
Figures illustrate differences in the time course for these two different
groups as well as a differential effect on paired-pulse facilitation in the
two groups.
Curve fitting on the time course of heterosynaptic depression was
utilized to estimate the time constants of these curves. The curves
showing average time course were fitted with equations representing
either the baseline level minus a single exponential decay, or the
baseline level minus a dual exponential time course. These time
constant parameters were obtained using the curve fitting function in
DeltaGraph 3.0.
Histology
At the end of the experiments, recording and stimulation sites were
marked by passage of a positive current of 5 mA for 2.5 s; this
produced a lesion and deposited iron from the electrode. After recording, rats were deeply anesthetized and perfused transcardially with
physiological saline followed by 250 ml of 10% formalin, with 10 ml
glacial acetic acid and 10 g of potassium ferrocyanide (to achieve a
Prussian Blue reaction with iron deposits). After removal, brains were
stored in 10% formalin/20% sucrose for 1–2 wk before sectioning.
Recording and stimulation sites were later localized on the basis of
both lesions and blue electrode marks after staining with Neutral Red.
87 • MARCH 2002 •
www.jn.org
TIME COURSE OF GABAB PRESYNAPTIC INHIBITION
1199
RESULTS
Heterosynaptic depression
Stimulation of the ipsilateral hippocampus caused heterosynaptic depression of potentials elicited by stimulation of
the contralateral hippocampus. An example of this heterosynaptic depression is illustrated in Fig. 2. This figure shows the
baseline potentials induced by contralateral stimulation 10 s
before the ipsilateral stimulation pulse. Repeating this same
pair of stimulus pulses after ipsilateral stimulation induced
synaptic potentials with much smaller initial negative slopes,
as seen in Fig. 2. As described in the INTRODUCTION and Fig. 1,
this heterosynaptic depression is theoretically due to the release
of GABA as a result of ipsilateral stimulation activating
GABAB receptors on synapses mediating contralateral evoked
synaptic potentials.
The term heterosynaptic depression specifically refers to
presynaptic inhibition caused at one set of synapses through
activation of a separate set of inputs to a region. To test
whether the decrease in synaptic potentials observed here resulted from heterosynaptic depression, experiments analyzed
whether depression only occurred with crossed ipsilateralcontralateral stimulation but not with ipsilateral-ipsilateral
stimulation or contralateral-contralateral stimulation. In all the
experiments described here, before heterosynaptic depression
was measured, pairs of stimuli on the same pathway were used
to test whether only paired-pulse facilitation appeared on each
pathway (a homosynaptic effect). Paired-pulse facilitation was
measured by comparing the negative slope of the second
synaptic potential to the negative slope of the first synaptic
potential induced 40 ms previously. Experiments were only
performed with electrode locations in which contralateral-contralateral pairs consistently induced paired-pulse facilitation (as
shown in Fig. 2) and ipsilateral-contralateral pairs did not. In
addition, experiments were only performed if ipsilateral-ipsilateral pairs also induced homosynaptic paired-pulse facilitation. This ruled out the possibility that depression arose from
synaptic vesicle depletion—a homosynaptic mechanism that
could cause depression at a number of synapses, particularly in
the neocortex. The depression was only observed with ipsilateral-contralateral stimulation—that is, if two different pathways were stimulated. If there was any overlap in the synapses
activated by ipsilateral and contralateral stimulation, it would
not contribute to heterosynaptic depression because these overlapping synapses would undergo facilitation.
Time course of heterosynaptic depression
Once heterosynaptic depression was obtained with ipsilateral-contralateral stimulation, the time course of this heterosynaptic depression was measured by varying the interval
between the ipsilateral conditioning pulse and the contralateral
test pulse between 30 and 1,500 ms. An example of the time
course of heterosynaptic depression in a single experiment is
shown in Fig. 3A. The magnitude of the initial negative slope
of potentials elicited by contralateral stimulation is plotted for
intervals ranging from 30 to 1,000 ms, demonstrating a rapid
early heterosynaptic depression that starts out strong and decays back to baseline (100%) in about 250 ms. The traces in
Fig. 3B show averaged evoked synaptic potentials at different
time delays in this example animal, clearly illustrating the
J Neurophysiol • VOL
FIG. 3. A: example of the time course of heterosynaptic depression appearing in rats showing weak initial paired-pulse facilitation. The initial slope of
potentials elicited by contralateral stimulation is plotted for intervals ranging
from 30 to 1,000 ms, demonstrating that heterosynaptic depression decays over
about 250 ms in rats starting with weak initial facilitation in response to
contralateral paired pulses. B: these traces shows the averaged potentials
elicited by the 1st pulse of the contralateral pulse pairs for a number of
different time points, including the baseline stimulation 10 s before the
ipsilateral pulse (control) and several different time delays after the ipsilateral
pulse. C: average time course of heterosynaptic depression in rats showing
weak initial paired-pulse facilitation. For each experiment, the initial slope of
averaged test potentials at different intervals was calculated and then normalized relative to the slope of the 1st baseline potential (baseline slope ⫽ 100%).
The mean for each rat was then combined with other available data at that time
point from other rats, giving an overall mean and a SE (plotted as error bars).
The number of recordings at each time point are represented by different
symbols (■ ⫽ 2 rats, Œ ⫽ 2 different rats, F ⫽ 4 rats). This illustrates a
consistent time course of heterosynaptic depression in rats that started with a
magnitude of paired-pulse facilitation less than 150% of control.
strength of heterosynaptic depression in this preparation. The
positive component at the start of the potential could reflect a
passive source induced by a more distant excitatory sink in the
molecular layer of the dentate gyrus. We feel this positive
component has been unmasked by the reduction in the local
potential rather than the passive source causing the entire change
in the local potential. This rat showed weak initial paired pulse
facilitation (less than 150% of control), which increased significantly during heterosynaptic depression (see Fig. 5).
Figure 3C shows average time course of heterosynaptic
87 • MARCH 2002 •
www.jn.org
1200
B. J. MOLYNEAUX AND M. E. HASSELMO
depression across several experiments. For each experiment,
the mean initial negative slope of test potentials at different
intervals was calculated and then normalized relative to the
initial negative slope of the first baseline potential (baseline
slope ⫽ 100%). The mean for each rat was then combined with
other available data at that time point from other rats with weak
initial paired pulse facilitation (less than 150% of control),
giving an overall mean and SE (plotted as error bars). The
number of recordings at each time point are represented by
different symbols. Data points were obtained at different time
delays in different rats, but they all fall on the same decay
curve in this figure. Using curve fitting with a single exponential, this data had a decay time constant of 141.5. If modeled
with a dual exponential (which would more accurately represent the properties of diffusion of GABA released in response
to the ipsilateral pulse), the best estimate was for a rising time
constant of 13.5 ms and a decay constant of 140.1 ms. The
weak initial paired-pulse facilitation in all these rats increased
significantly during the period of heterosynaptic depression
(see Fig. 5).
As described in METHODS, different properties appeared in
rats in which the initial paired-pulse facilitation in response to
contralateral stimulation was larger. An example of the faster
time course of heterosynaptic depression in rats with strong
initial facilitation (more than 180% of control) is shown in Fig.
4A. Figure 4B shows averaged evoked potentials for different
time delays. Figure 4C shows the average time course of
heterosynaptic depression in all rats showing strong initial
paired-pulse facilitation (more than 180% of control). This
illustrates a consistently faster time course of heterosynaptic
depression in experiments that started with a stronger magnitude of paired-pulse facilitation. In these experiments, the
heterosynaptic depression was somewhat stronger on average
and decayed back to baseline more rapidly, reaching values at
100% of baseline in less than 180 ms. The rapid time course
seen in the example in Fig. 4A appears consistently across
seven different experiments with strong initial facilitation. If
the curve was modeled with a single exponential decay, the
time constant is estimated as 72.3 ms. If the curve was modeled
as a dual exponential, then the best estimate would have a
rising time constant of 16.1 ms and a decay time constant of
59.3.
Anatomical localization of the recording sites for these two
different groups of rats showed that recordings with weak
facilitation appeared to be consistently more distal in s. radiatum, whereas at least some of the recordings with strong
facilitation appeared to be more proximal to s. pyramidale.
However, the size of the lesions marking electrode location
prevented an accurate quantitative analysis of these differences. It is very likely that these differences are due to electrode placement in specific experiments rather than differences
in the physiology of individual rats.
Interaction of heterosynaptic depression and paired-pulse
facilitation
The changes in paired-pulse facilitation during the period of
heterosynaptic depression were also examined. As noted in the
preceding text and in Fig. 2, in all experiments, presentation of
homosynaptic pulse pairs to each pathway was used to test
whether paired-pulse facilitation was present before testing for
J Neurophysiol • VOL
FIG. 4. A: example of the time course of heterosynaptic depression in rats
showing strong initial paired-pulse facilitation. The initial slope of potentials
elicited by contralateral stimulation is plotted for intervals ranging from 30 to
1,000 ms, demonstrating a rapid early heterosynaptic paired-pulse depression
that has returned to baseline within about 180 ms. B: these traces show the
averaged potentials elicited by the 1st pulse of the contralateral pulse pairs for
a number of different time points, including the baseline stimulation 10 s
before the ipsilateral pulse (control) and several different time delays after the
ipsilateral pulse. Pulses show a strong decrease in initial slope and overall
magnitude during heterosynaptic depression. C: average time course of heterosynaptic depression across several experiments. The number of recordings
at each time point are represented by different symbols (■ ⫽ 6 rats, ● ⫽ 7 rats).
These data illustrate a consistent time course across all rats with strong initial
paired pulse facilitation to contralateral stimulation (facilitation more than
180% of control).
heterosynaptic depression. The additional analysis described
here was utilized to determine if there was a change in magnitude of this previously measured contralateral paired-pulse
facilitation during heterosynaptic depression. The magnitude
of paired-pulse facilitation is measured by comparing the second pulse of the contralateral paired-pulse stimulation with the
corresponding first pulse at that interval. For example, the
second pulse of a pair consisting of one pulse at 30- and a
second pulse at 70-ms delays was compared with the first pulse
of a pair at presented at 70 and 110 ms. In the figures shown
here, the magnitude of paired-pulse facilitation at different
time delays after ipsilateral stimulation (䊐) is plotted relative to
the paired-pulse facilitation observed in the baseline pulses
10 s before ipsilateral stimulation (E).
87 • MARCH 2002 •
www.jn.org
TIME COURSE OF GABAB PRESYNAPTIC INHIBITION
1201
Figure 5A shows an example of the change in paired-pulse
facilitation during and after heterosynaptic depression in an
experiment starting with weak paired-pulse facilitation (around
140% of control in this example). The experiment shows a
large increase in paired-pulse facilitation during the period of
heterosynaptic depression (▫) as compared with the facilitation
obtained during the baseline period preceding each test interval
by 10 s (E). Figure 5B shows the change in paired-pulse
facilitation across the full set of experiments in which facilitation started out weak. There was a consistent increase in
paired-pulse facilitation across this set of experiments. The
best fit of a dual exponential to this equation was obtained with
delay to onset of 27.2 ms, and equal rising and falling time
constants of 68.7 ms. Previous experiments assume that presynaptic inhibition should be accompanied by an increase in
paired-pulse facilitation (Dittman and Regehr 1997; Valentino
and Dingledine 1981; Zucker 1989). If this is true, the time
course of increased facilitation seen in these experiments might
give the most accurate estimate of presynaptic inhibition.
Figure 6A shows the change in paired-pulse facilitation
during and after heterosynaptic depression in a rat starting with
strong paired-pulse facilitation. This rat showed a decrease in
FIG. 6. A: example of the change in paired-pulse facilitation during and
after heterosynaptic depression in preparations starting with strong pairedpulse facilitation. 䊐, average paired-pulse facilitation in this experiment at
different intervals after ipsilateral stimulation; E, the average paired-pulse
facilitation obtained with baseline stimulation 10 s before each of the 䊐 plotted
here. Note that these all show very strong initial baseline facilitation (around
220% of control). This rat showed a decrease in paired-pulse facilitation during
heterosynaptic depression. B: average change in paired-pulse facilitation during and after heterosynaptic depression in preparations starting with strong
paired-pulse facilitation. 䊐, data points averaged over 6 rats; E, points averaged
over 7 rats. The change in facilitation is consistent across all rats with strong
initial facilitation.
FIG. 5. A: example of the change in paired-pulse facilitation during and
after heterosynaptic depression in a rat starting with weak paired pulse facilitation (about 140% of control in this example). These rats showed a clear
increase in magnitude of paired-pulse facilitation during heterosynaptic depression. 䊐, the paired-pulse facilitation at different intervals after ispilateral
stimulation; E, the baseline control measurements given 10 s before each of the
paired pulses plotted with 䊐. (Note that baseline measurements involved paired
pulses that are plotted according to the time interval between the ipsilateral
pulse and the later contralateral paired pulse.) Baseline measurements are also
shown for long interval paired pulses where faciliation measurements were not
obtained. B: average change in paired-pulse facilitation during and after
heterosynaptic depression in preparations starting with weak paired-pulse
facilitation (less than 150% of control). The rats starting with weak pairedpulse facilitation show a consistent increase in paired-pulse facilitation during
the period of heterosynaptic depression from 0 to 250 ms (note the difference
in time scale from Figs. 3 and 4). Symbols represent different numbers of data
points at different intervals (䊐 ⫽ 2 rats, ‚ ⫽ 2 different rats, E ⫽ 4 rats).
J Neurophysiol • VOL
paired-pulse facilitation during heterosynaptic depression with
a time course of change similar to the time course observed for
heterosynaptic depression. However, note that even though the
magnitude was decreased, contralateral stimulation pairs still
exhibited paired-pulse facilitation, again suggesting that the
heterosynaptic depression is not due to homosynaptic effects.
Figure 6B shows the average change in magnitude of pairedpulse facilitation across all the experiments starting with strong
paired-pulse facilitation. For these experiments, the facilitation
was consistently reduced from the initial high baseline during
heterosynaptic depression with a similar time course across
experiments. This may suggest a postsynaptic contribution to
heterosynaptic depression in these experiments.
Pharmacology
To determine whether the observed heterosynaptic pairedpulse depression was due to activation of presynaptic GABAB
receptors, we utilized local infusion of pharmacological agents
near the field potential recording site. To test this technique, we
performed local infusion of the glutamatergic AMPA receptor
antagonist CNQX in a single experiment, expecting to see a
decrease in the magnitude of the excitatory field potentials.
This infusion of CNQX caused the field potentials to decrease
in size, as shown in Fig. 7. A complete blockade of potentials
was not expected, as infusion was done at relatively low
87 • MARCH 2002 •
www.jn.org
1202
B. J. MOLYNEAUX AND M. E. HASSELMO
FIG. 7. Evaluation of the effectiveness of local infusion by testing effects of
infusion of the AMPA receptor blocker 6-cyano-7-nitroquinoxalene-2,3-dione
(CNQX) on field potentials recorded in s. radiatum of region CA1 during
stimulation of the contralateral hippocampus. Top: slope of synaptic potentials
plotted against time before, during, and after local infusion of CNQX through
the cannula near recording site. Bottom: averages of field potentials recorded
before infusion (pre), during infusion (CNQX), and after infusion (wash).
concentrations to prevent nonspecific effects. After a 75-min
wash period, the potentials had returned to their preinfusion
magnitude.
Next, the effect of the GABAB antagonist CGP55845A
(Davies et al. 1993) on heterosynaptic depression was tested. In
these experiments, the time course of heterosynaptic depression was first recorded with a range of different delay intervals
between ipsilateral and contralateral pulses. Then to observe
the effect of pharmacological infusion, the level of depression
was monitored with consistent alternating use of contralateral
pulse pairs starting at only two intervals (40 or 80 ms) after the
ipsilateral pulse. The magnitude of heterosynaptic depression
at these time intervals remained consistent across a 35-min
baseline period, after which infusion of CGP55845A was initiated. The infusion of CGP 55845A strongly reduced the
depression in four of four animals showing strong initial facilitation and also strongly reduced the depression in the one
animal tested with weak initial facilitation. Thus the pharmacological effect was consistent in every animal tested. Approximately 10 min after the CGP 55845A infusion was initiated,
the level of depression was greatly reduced. These data suggest
that a significant portion of the heterosynaptic depression resulted from activation of GABAB receptors.
Figure 8A shows the effect of the GABAB receptor antagonist on heterosynaptic depression in the one case of weak initial
facilitation in which pharmacological infusion was tested. The
initial negative slope of potentials is plotted relative to the
baseline. Before infusion, the heterosynaptic depression at 40
ms maintained a consistent level of about 35% reduction from
baseline. After about 10 min of antagonist infusion, the level of
heterosynaptic depression was greatly reduced, returning the
initial negative slope of potentials to a magnitude close to the
baseline magnitude (observed 10 s before presentation of the
ipsilateral pulse). In this experiment, the increase in facilitation
caused by heterosynaptic depression was reduced when the
heterosynaptic depression was reduced by the antagonist. The
initial negative slope of the baseline potentials did increase
during infusion of antagonists, possibly due to loss of tonic
endogenous activation of GABAB receptors, but the decrease
in magnitude of depression was greater than would result from
the increase in baseline alone. Figure 8B shows the effect of the
GABAB receptor antagonist on the magnitude of heterosynaptic depression at the 40-ms interval in the four experiments
starting with strong facilitation. Before drug infusion, the average depression at 40 ms remained steady at a mean of about
60% below baseline. Drug infusion consistently reduced the magnitude of depression to an average magnitude of about 30%.
Infusion of the GABAB receptor antagonist CGP55845A
had effects on paired-pulse facilitation consistent with the
FIG. 8. A: the effect of local infusion of the GABAB receptor
antagonist CGP 55845A on the magnitude of heterosynaptic depression for contralateral evoked potentials following ipsilateral
stimulation by 40 ms in the experiment starting with weak facilitation in which drug infusion was performed. B: the effect of local
infusion of the GABAB receptor antagonist CGP55845A on the
magnitude of heterosynaptic depression at 40 ms in the experiments starting with strong facilitation in which pharmacological
infusion was performed (n ⫽ 4).
J Neurophysiol • VOL
87 • MARCH 2002 •
www.jn.org
TIME COURSE OF GABAB PRESYNAPTIC INHIBITION
notion that this antagonist reduces the magnitude of heterosynaptic depression. As described in the preceding text, presynaptic inhibition during heterosynaptic depression (after ipsilateral stimulation) should reduce the size of the first negative
potential in each pair of potentials, leaving a larger store of
unreleased transmitter and allowing greater paired-pulse facilitation (Dittman and Regehr 1997; Valentino and Dingledine
1981; Zucker 1989). Blockade of this presynaptic inhibition by
the antagonist should increase the size of the first potential in
each pair after ipsilateral stimulation, resulting in a reduction in
the amount of paired-pulse facilitation.
As shown in Fig. 9, A and B, the GABAB receptor antagonist
caused a reduction in the amount of facilitation of test pulses
(after ipsilateral stimulation) relative to the facilitation of baseline potentials (before ipsilateral stimulation). The increase in
the amount of facilitation of baseline pulse pairs (before ipsilateral stimulation) could result from blockade of the GABABmediated presynaptic inhibition of inhibitory synaptic transmission induced by the first pulse in each pair. As shown in
Fig. 9B, average data from four cases starting with strong
facilitation shows that there was a significant absolute reduction in the facilitation of test potentials. The reduction in
facilitation shown in this figure is consistent with the idea that
the GABAB receptor antagonist blocks presynaptic inhibition,
allowing more transmitter release in response to the first pulse,
1203
leaving less transmitter available for the second pulse, and
thereby resulting in less paired-pulse facilitation.
DISCUSSION
These in vivo data demonstrate that heterosynaptic depression in region CA1 of anesthetized rats has a time course which
is more rapid than was previously inferred from in vitro data.
This heterosynaptic depression was significantly blocked by
local infusion of the GABAB receptor antagonist CGP 55845A,
suggesting that presynaptic inhibition by GABAB receptors
contributes to a significant component of this heterosynaptic
depression, as proposed for heterosynaptic depression in brain
slice preparations of hippocampal region CA1 (Isaacson et al.
1993) and cerebellum (Dittman and Regehr 1997).
The time course of heterosynaptic depression observed here
appears sufficiently rapid to underlie significant changes in
strength of synaptic transmission within each cycle of the theta
rhythm. This evidence supports the possibility that presynaptic
inhibition by GABAB receptors could contribute to the observed changes in magnitude of evoked synaptic potentials at
different phases of the theta cycle (Rao et al. 1998; Wyble et al.
2000). In addition, these effects could contribute to the change
in magnitude of evoked population spikes at different phases of
the theta cycle (Buzsaki et al. 1981; Rudell and Fox 1984;
Rudell et al. 1980).
Relation to heterosynaptic depression in slice preparations
FIG. 9. Effect of infusion of the GABAB antagonist CGP 55845A on the
magnitude of paired-pulse facilitation. A: effect of GABAB antagonist in the
experiment starting with weak facilitation. E, after antagonist infusion there is
an increase in facilitation of the baseline pulse pair (stimulated before ipsilateral stimulation). 䊐, facilitation of the test pulse pair (after ipsilateral stimulation) shows transient changes followed by stabilization at the same level as
baseline. This reduction in the relative amount of faciliation is consistent with
the reduction in heterosynaptic depression caused by the antagonist. B: effect
of antagonist in 4 experiments starting with strong facilitation. E, an increase
in average facilitation in baseline pulse pairs (before ipsilateral stimulation)
consistent with A. 䊐, the magnitude of facilitation of the test pulse pair (after
ipsilateral stimulation) was significantly reduced after antagonist. Again, this
suggests that the blockade of presynaptic inhibition by CGP55845A allows
more transmission for the 1st pulse and thereby decreases the amount of
facilitation possible with the 2nd pulse.
J Neurophysiol • VOL
The time course of heterosynaptic depression observed was
significantly more rapid than the time course observed in slice
experiments. In those experiments, the time course of decay of
heterosynaptic depression was around 1 s, whereas here the
depression decayed to baseline in less than 300 ms.
These differences could arise from different factors including the use of nonphysiological temperatures in most slice
experiments and the change in amount of GABA reuptake and
in intracellular metabolic processes in slice preparations. In
slice experiments (Dittman and Regehr 1996), the Q10 of 3.3
reported for changes in the time course of heterosynaptic
depression at different temperatures is consistent with a much
faster time course at the temperatures maintained in these
experiments (37°C), although the exact values extrapolated for
physiological temperatures are still somewhat slower than
those observed here. They reported a decay time constant of
680 ms at 32°, which for a Q10 of 3.3 would result in a decay
time constant of 412 ms at 37°. Blockade of GABA reuptake
has been demonstrated to enhance heterosynaptic depression in
the slice preparation (Isaacson et al. 1993; Mitchell and Silver
2000). Stronger reuptake in vivo could significantly speed the
decay of extracellular activation of GABAB receptors, just as it
influences the speed of decay of GABAA currents (Mody et al.
1994; Rossi and Hamann 1998) and glutamatergic currents
(Asztely et al. 1997).
Relation to functional changes during theta
These results suggest that phasic changes in the magnitude
of excitatory postsynaptic potentials (EPSPs) and populations
spikes during theta could at least partly result from changes in
activation of presynaptic GABAB receptors. Computational
modeling suggests that such phasic changes in strength of
87 • MARCH 2002 •
www.jn.org
1204
B. J. MOLYNEAUX AND M. E. HASSELMO
synaptic transmission could allow separate phases of encoding
and retrieval in the hippocampal formation (Hasselmo et al.
1996, 2001; Wallenstein and Hasselmo 1997). Models of hippocampal function demonstrate that effective encoding of new
information requires weakening of synaptic transmission at
connections mediating retrieval—including the Schaffer collaterals studied here (Hasselmo et al. 1996, 2001). Weak transmission prevents the retrieval of previously encoded information from being re-encoded and interfering with new learning.
The GABAergic presynaptic inhibition of transmission at the
Schaffer collaterals described here could provide a mechanism
for phasically weakening retrieval activity. During a separate
retrieval phase, the rapid decay of this presynaptic inhibition
could allow stronger transmission at the Schaffer collaterals,
allowing retrieval activity to activate region CA1 neurons and
cause retrieval output from the hippocampus (Hasselmo et al.
2001).
Modeling (Hasselmo et al. 2001) can account for behavioral
data showing that lesions of the medial septum or the fornix,
which block theta rhythm generation in the hippocampus,
cause strong impairments in spatial reversal tasks (M’Harzi et
al. 1987; Whishaw and Tomie 1997) and delayed alternation
tasks (Numan and Quaranta 1990; Numan et al. 1995). In these
tasks, animals must distinguish the retrieval of previously
encoded information from the encoding of new sensory information. Computational modeling demonstrates how phasic
changes in the strength of intrinsic transmission could enhance
this distinction by reducing the influence of retrieval activity
during the encoding of new information (Hasselmo et al. 1996,
2001; Sohal and Hasselmo 1998a,b; Wallenstein and Hasselmo
1997). Loss of this phasic modulation due to lesions allows
retrieval of previous memories for reward location to occur
during new encoding, even if the reward has been moved to a
different location. This prevents the extinction of initial learned
associations and greatly slows the learning of a reversal (Hasselmo et al. 2001), consistent with behavioral data. Modeling
also shows that phasic changes allow effective retrieval of
weak associations in a network with other associations of
similar or greater strength (Sohal and Hasselmo 1998a,b). This
modeling provides a circuit level mechanism for previously
proposed functions of theta, including selection of novel and
significant stimuli for encoding in the face of irrelevant stimuli
(Oddie and Bland 1998; Vinogradova 1995).
The separation of encoding and retrieval phases requires that
synaptic modification should be strongest at the time of weakest synaptic transmission. This requirement is supported by
evidence for the best induction of LTP at the peak of the local
theta when transmission is weakest (Hasselmo et al. 2001;
Holscher et al. 1997; Huerta and Lisman 1993; Pavlides et al.
1988). In fact, the activation of GABAB receptors appears to
simultaneously decrease synaptic transmission and enhance
LTP (Mott and Lewis 1991, 1994).
Modeling also demonstrates that phasic changes in magnitude of synaptic transmission during theta could contribute to
the theta phase precession effect (O’Keefe and Recce 1993;
Skaggs et al. 1996). During theta phase precession, the spiking
of place cells appears later in a theta cycle when a rat first
enters the place field and then moves to earlier phases. This
effect can be obtained in the model if GABAB presynaptic
inhibition is strong at early phases, ensuring that place cells are
only activated by entorhinal cortical input bearing information
J Neurophysiol • VOL
about external cues (Wallenstein and Hasselmo 1997). As the
presynaptic inhibition becomes weaker at late phases of theta,
this allows previously modified excitatory synapses to cause
associative activation of place cells further along in a familiar
pathway, causing late phase firing to be strong as a rat first
enters the place field (Wallenstein and Hasselmo 1997).
We appreciate the help of B. P. Wyble and C. Linster in setting up the
techniques utilized here.
This work was supported by National Institute of Mental Health Grants
MH-60013 and MH-61492, National Science Foundation Grant IBN9996177,
and a grant from the Human Frontier Science Program.
Present address of B. J. Molyneaux: Program in Neuroscience, Harvard
Medical School, Boston, MA 02115.
REFERENCES
AULT B AND NADLER JV. Baclofen selectively inhibits transmission at synapses
made by axons of CA3 pyramidal cells in the hippocampal slice. J Pharm
Exp Ther 223: 291–297, 1982.
ASZTELY F, ERDEMLI G, AND KULLMANN DM. Extrasynaptic glutamate spillover in the hippocampus: dependence on temperature and the role of active
glutamate uptake. Neuron 19: 281–293, 1997.
BUZSAKI G, GRASTYAN E, CZOPF J, KELLENYI L, AND PROHASKA O. Changes in
neuronal transmission in the rat hippocampus during behavior. Brain Res
225: 235–247, 1981.
BUZSAKI G, LEUNG LW, AND VANDERWOLF CH. Cellular bases of hippocampal
EEG in the behaving rat. Brain Res 287: 139 –171, 1983.
CHROBAK JJ AND BUZSAKI G. Selective activation of deep layer (V-VI) retrohippocampal cortical neurons during hippocampal sharp waves in the behaving rat. J Neurosci 14: 6160 – 6170, 1994.
COLBERT CM AND LEVY WB. Electrophysiological and pharmacological characterization of perforant path synapses in CA1: mediation by glutamate
receptors. J Neurophysiol 68: 1–7, 1992.
DAVIES CH, POZZA MF, AND COLLINGRIDGE GL. CGP 55845A: a potent
antagonist of GABAB receptors in the CA1 region of the hippocampus.
Neuropsychopharmacology 32: 1071–1073, 1993.
DITTMAN JS AND REGEHR WG. Mechanism and kinetics of heterosynaptic
depression at a cerebellar synapse. J Neurosci 17: 9048 –9059, 1997.
FOX SE, WOLFSON S, AND RANCK JBJ. Hippocampal theta rhythm and the
firing of neurons in walking and urethane anesthetized rats. Exp Brain Res
62: 495–508, 1986.
GREEN JD AND ARDUINI AA. Hippocampal electrical activity and arousal.
J Neurophysiol 17: 533–557, 1954.
HASSELMO ME, BODELON C, AND WYBLE BP. A proposed function for hippocampal theta rhythm: separate phases of encoding and retrieval enhance
reversal of prior learning. Neural Comput 14: 792– 818, 2002.
HASSELMO ME, WYBLE BP, AND WALLENSTEIN GV. Encoding and retrieval of
episodic memories: role of cholinergic and GABAergic modulation in the
hippocampus. Hippocampus 6: 693–708, 1996.
HOLSCHER C, ANWYL R, AND ROWAN MJ. Stimulation on the positive phase of
hippocampal theta rhythm induces long-term potentiation that can be depotentiated by stimulation on the negative phase in area CA1 in vivo. J Neurosci 17: 6470 – 6477, 1997.
HUERTA PT AND LISMAN JE. Heightened synaptic plasticity of hippocampal
CA1 neurons during a cholinergically induced rhythmic state. Nature 364:
723–725, 1993.
ISAACSON JS, SOLIS JM, AND NICOLL RA. Local and diffuse synaptic actions of
GABA in the hippocampus. Neuron 10: 165–175, 1993.
LILLIE LE, TEMPLE NJ, AND FLORENCE LZ. Reference values for young normal
Sprague-Dawley rats: weight gain, hematology and clinical chemistry. Hum
Exp Toxicol 15: 612– 616, 1996.
LINSTER C, WYBLE BP, AND HASSELMO ME. Electrical stimulation of the
horizontal limb of the diagonal band of Broca modulates population EPSPs
in piriform cortex. J Neurophysiol 81: 2737–2742, 1999.
M’HARZI M, PALACIOS A, MONMAUR P, WILLIG F, HOUCINE O, AND DELACOUR
J. Effects of selective lesions of fimbria-fornix on learning set in the rat.
Physiol Behav 40: 181–188, 1987.
MITCHELL SJ AND SILVER RA. GABA spillover from single inhibitory axons
suppresses low-frequency excitatory transmission at the cerebellar glomerulus. J Neurosci 20: 8651– 8658, 2000.
MODY I, DEKONINCK Y, OTIS TS, AND SOLTESZ I. Bridging the cleft at GABA
synapses in the brain. Trends Neurosci 17: 517–525, 1994.
87 • MARCH 2002 •
www.jn.org
TIME COURSE OF GABAB PRESYNAPTIC INHIBITION
MOTT DD AND LEWIS DV. Facilitation of the induction of long-term potentiation by GABAB receptors. Science 252: 1718 –1720, 1991.
MOTT DD AND LEWIS DV. The pharmacology and function of central GABAB
receptors. Int Rev Neurobiol 36: 97–223, 1994.
NUMAN R, FELONEY MP, PHAM KH, AND TIEBER LM. Effects of medial septal
lesions on an operant go/no-go delayed response alternation task in rats.
Physiol Behav 58: 1263–1271, 1995.
NUMAN R AND QUARANTA JR JR. Effects of medial septal lesions on operant
delayed alternation in rats. Brain Res 531: 232–241, 1990.
ODDIE SD AND BLAND BH. Hippocampal formation theta activity and movement selection. Neurosci Biobehav Rev 22: 221–231, 1998.
O’KEEFE J AND RECCE ML. Phase relationship between hippocampal place
units and the EEG theta rhythm. Hippocampus 3: 317–330, 1993.
OTIS TS, DEKONINCK Y, AND MODY I. Characterization of synaptically elicited
GABA(B) responses using patch-clamp recordings in rat hippocampal
slices. J Physiol (Lond) 463: 391– 407, 1993.
OTTO T, EICHENBAUM H, WIENER SI, AND WIBLE CG. Learning-related patterns
of CA1 spike trains parallel stimulation parameters optimal for inducing
hippocampal long-term potentiation. Hippocampus 1: 181–192, 1991.
PAULSEN O AND MOSER EI. A model of hippocampal memory encoding and
retrieval: GABAergic control of synaptic plasticity. Trends Neurosci 21:
273–278, 1998.
PAVLIDES C, GREENSTEIN Y, GRUDMAN M, AND WINSON J. Long-term potentiation in the dentate gyrus is induced preferentially on the positive phase of
theta rhythm. Brain Res 439: 383–387, 1988.
PAXINOS G AND WATSON C. The Rat Brain in Stereotaxic Coordinates. San
Diego, CA: Academic, 1986.
PFRIEGER FW, GOTTMAN, AND LUX HD. Kinetics of GABAB receptor-mediated inhibition of calcium currents and excitatory synaptic transmission in
hippocampal neurons in vitro. Neuron 12: 97–107, 1994.
RAO G, ORR G, BARNES CA, AND MCNAUGHTON BL. Theta rhythm modulation
of CA1 evoked potential size during locomotion. Soc Neurosci Abstr 24:
933, 1998.
RICK JT AND MILGRAM NW. Instability of dentate gyrus field potentials in
awake and anesthetized rats. Hippocampus 9: 333–339, 1999.
ROSSI DJ AND HAMANN M. Spillover-mediated transmission at inhibitory
synapses promoted by high affinity a6 subunit GABAA receptors and
glomerular geometry. Neuron 20: 783–795, 1998.
RUDELL AP AND FOX SE. Hippocampal excitability related to the phase of the
theta rhythm in urethanized rats. Brain Res 294: 350 –353, 1984.
RUDELL AP, FOX SE, AND RANCK JBJ. Hippocampal excitability phase-locked
to the theta rhythm in walking rats. Exp Neurol 68: 87–96, 1980.
J Neurophysiol • VOL
1205
SKAGGS WE, MCNAUGHTON BL, WILSON MA, AND BARNES CA. Theta phase
precession in hippocampal neuronal populations and the compression of
temporal sequences. Hippocampus 6: 149 –172, 1996.
SOHAL VS AND HASSELMO ME. Changes in GABAB modulation during a theta
cycle may be analogous to the fall of temperature during annealing. Neural
Comput 10: 889 –902, 1998a.
SOHAL VS AND HASSELMO ME. GABAB modulation improves sequence disambiguation in computational models of hippocampal region CA3. Hippocampus 8: 171–193, 1998b.
STEWART M AND FOX SE. Do septal neurons pace the hippocampal theta
rhythm? Trends Neurosci 13: 163–168, 1990.
TOTH K, FREUND TF, AND MILES R. Disinhibition of rat hippocampal pyramidal
cells by GABAergic afferent from the septum. J Physiol (Lond) 500:
463– 474, 1997.
VALENTINO RJ AND DINGLEDINE R. Presynaptic inhibitory effect of acetylcholine in the hippocampus. J Neurosci 1: 784 –792, 1981.
VINOGRADOVA OS. Expression, control and probable functional significance of
the neuronal theta rhythm. Prog Neurobiol 45: 523–583, 1995.
VOGT KE AND NICOLL RA. Glutamate and g-aminobutyric acid mediate a
heterosynaptic depression at mossy fiber synapses in the hippocampus. Proc
Natl Acad Sci USA 96: 1118 –1122, 1999.
VOGT KE AND REGEHR WG. Cholinergic modulation of excitatory synaptic
transmission in the CA3 area of the hippocampus. J Neurosci 21: 75– 83,
2001.
WALLENSTEIN GV, EICHENBAUM HB, AND HASSELMO ME. The hippocampus as
an associator of discontiguous events. Trends Neurosci 21: 317–323, 1998.
WALLENSTEIN GV AND HASSELMO ME. GABAergic modulation of hippocampal population activity: sequence learning, place field development and the
phase precession effect. J Neurophysiol 78: 393– 408, 1997.
WHISHAW IQ AND TOMIE JA. Perseveration on place reversals in spatial
swimming pool tasks: further evidence for place learning in hippocampal
rats. Hippocampus 7: 361–370, 1997.
WYBLE BP, LINSTER C, AND HASSELMO ME. Evoked synaptic potential size
depends on phase of theta rhythm in rat hippocampus. Soc Neurosci Abstr
23: 508, 1997.
WYBLE BP, LINSTER C, AND HASSELMO ME. Size of CA1-evoked synaptic
potentials is related to theta rhythm phase in rat hippocampus. J Neurophysiol 83: 2138 –2144, 2000.
ZUCKER RS. Short-term synaptic plasticity. Annu Rev Neurosci 12: 13–31,
1989.
87 • MARCH 2002 •
www.jn.org
Download